Optimizing Test Case Reduction Using Particle Swarm Optimization

Testing is one of the most important phases of software development, ensuring that the system functions efficiently, performs its intended tasks, and remains reliable. As the size of the test set grows polynomially (N²), the cost, redundancy and executing time also increase. Classical test case reduction techniques (heuristic and greedy approaches) usually cannot strike the right balance between the capability to detect errors and code coverage in minimizing test cases. In order to tackle this problem, we first construct a PSObased testing case reduction method to intelligently select the optimal test case subset. The PSO algorithm considers the test case selection as an optimization problem and uses particles to represent candidate subsets. The objective of the fitness function is to maximize the defect finding rate, code coverage and minimize execution time. With the update of positions with respect to global and individual optima, particles can conduct detailed search and decision. Experimental results show that our approach achieves high defect detection and coverage by greatly reducing the total number of tests. This renders the method particularly well suited to large software testing with less overhead, a more rapid execution, and a higher effectiveness.